EVgo, Inc. (EVGO) Earnings Call Transcript & Summary

Thank you for standing by, and welcome to the EVgo DOE Loan Close Conference Call. [Operator Instructions] I'd now like to turn the call over to Heather Davis, Vice President of Investor Relations at EVgo. You may begin.

Hello, and welcome to EVgo's call to discuss our DOE loan. My name is Heather Davis, and I'm the Vice President of Investor Relations at EVgo. Joining me on today's call are Badar Khan, EVgo's Chief Executive Officer; and Paul Dobson, EVgo's Chief Financial Officer. Today, we will be discussing the $1.25 billion guaranteed loan facility EVgo received through the Department of Energy's Loan Program Office and its impact on our growth plan. Today's call is being webcast and can be accessed at the Investors section of our website at investors.evgo.com. The call will be archived and available there, along with the investor presentation, after the conclusion of this call. During the call, management will be making forward-looking statements that are subject to risks and uncertainties, including expectations about future performance. Factors that could cause actual results to differ materially from our expectations or details in our SEC filings, including in the Risk Factors section of our most recent annual report on Form 10-K and quarterly report on Form 10-Q. The company's SEC filings are available on the Investors section of our website. These forward-looking statements apply as of today, and we undertake no obligation to update these statements after the call. Also, please note that we will be referring to certain non-GAAP financial measures on this call. Information about these non-GAAP measures, including a reconciliation to the corresponding GAAP measures, can be found in the accompanying investor presentation available on the Investors section of our website. I'll turn the call over to Badar Khan, EVgo's CEO.

Thank you, Heather, and welcome to everyone joining us today. We are thrilled to announce EVgo close on a $1.25 billion guaranteed loan facility from the Department of Energy Loan Programs Office after the end of a roughly 18-month process. This loan will allow us to accelerate our mission of bringing convenient, reliable and affordable fast charging to all EV drivers across the United States. Emissions from transportation represent the largest source of emissions in the U.S., and that is why the work we do is so important. Our work is also a key ingredient to the long-term competitiveness and sustainability of the U.S. automotive industry. There is an unmistakable trend towards electrifying transportation across the globe that China is currently winning. Range anxiety with the lack of public charging infrastructure is often cited as one of the main barriers to faster EV adoption. In the U.S., we have around 100 electric vehicles for every public fast charger. In China, it's 50. Accelerating the buildout of public fast charging is critical to building driver's range confidence and EV demand in the U.S., which will allow U.S. automakers to scale up their EV businesses and become more competitive globally in an industry that currently provides over 1 million manufacturing jobs in the United States. As you know, EVgo is currently the third largest and one of the fastest-growing owner-operators of DC fast charging in the U.S. with a demonstrated track record of growth. Charging network throughput has grown nearly tenfold in the past 3 years, significantly higher than the growth in VIO. From customer capture to site development and construction, to products and services that build customer loyalty, we have built a growth engine across this entire cycle that is hard to replicate. Foundational to this growth engine are the many years experience we have been securing in our supply chain, marquee site host relationships, excellent relationships and advocacy efforts with governments and utilities, an innovative tech platform and our sizable OEM partnerships. One of EVgo's sources of competitive advantage honed from over a decade's worth of data about charging demand and behavior is a proprietary sophisticated network planning process that informs where we locate our chargers. We ingest an enormous amount of data, from EV adoption rates to forecasted sales, to density multifamily housing, to rideshare volumes, to electricity costs, to demand charges and availability of grants, all at a census block group level, which then tells us where to place chargers, how many and at what pace to generate double-digit returns. The result is a predictable, recurring and growing set of cash flows, much like many high-growth infrastructure asset classes that are well suited to project finance. We finalized an attractive and flexible loan with the DOE Loan Programs Office that ensures we have full funding to build approximately 7,500 charging stalls over 5 years without the need to raise equity to finance this build program. EVgo will develop charging stalls that will be transferred to a project finance SPV, which is an entity wholly owned by EVgo, at completion. EVgo will borrow up to 80% of CapEx and development costs that had incurred, cash flows generated within this borrower entity, but also reimbursed EVgo for most of the remaining 20% of CapEx and development costs incurred as well as ongoing expenses subject to reserve accounts and debt service coverage ratios. There are several components of the loan structure that are important to emphasize. First, EVgo will contribute a portion of its existing public network that provides day 1 collateral value and cash flows that will provide initial operating revenues to the SPV and whose operating performance will offer a margin of safety with respect to performance tests. As a result, the company is not expected to need to raise additional equity. Second, we receive reimbursements from the SPV in respect of project costs monthly. And together with an initial $75 million advance, this largely mitigates any working capital funding needs for the development of these stalls. Third, loan covers up to 100% of the gross CapEx and development cost per stall through monthly draws and reimbursements, which are expected to be received within a month or 2 of the stall becoming operational, again, minimizing any working capital needs. There are certain performance tests that need to be satisfied to receive these monthly draws that are effectively based on stall performance. There's considerable flexibility in these draw conditions where draws continue even in a very unlikely scenario of no further growth in throughput per stall from current levels. Fourth, after paying for direct expenses, excess cash flows generated by the project during both the development period and the operational period can be distributed back to EVgo once certain reserve accounts have been funded and after any debt service payments. There are certain performance tests that need to be satisfied for EVgo to be able to receive these distributions that are, again, effectively based on stall performance. We expect that there will be sufficient cash flows to start these distributions as early as Q1 of 2026, and we do not forecast any restrictions caused by the performance tests. Fifth, we have the ability to develop stalls faster or slower than planned if circumstances at the time compel us to do so. We may also build more stalls if we deliver improvements to gross CapEx per stall. The loan is sized based on building 7,500 stalls at a gross CapEx cost per stall largely in line with what we initially expected at the start of 2024. As we have said on prior earnings calls, we now expect a 6% reduction in gross CapEx per stall in vintage 2024 stalls, a 10% reduction in vintage 2025 stalls and are well underway in a program for next-generation architecture to deliver a 30% reduction in gross CapEx per stall for new stalls deployed on the second half of 2026. This lower capital cost per stall allows the company to build an estimated additional 1,600 stalls for the same loan amount, resulting in even higher adjusted EBITDA and cash flow by the end of the development period. Sixth, although over 40% of new stalls are expected to be in marginalized areas that have been overburdened by environmental impacts, this does not represent a requirement or condition of the loan but rather an assessment that these communities are aligned with our network plan that prioritizes rideshare and drivers more likely to live in multifamily housing. We also expect to continue to realize 30C tax credits for developing in these areas. As a result, our site selection parameters remain unchanged, and therefore, we expect our projected unit economics to remain unchanged. Seventh, this is a long-term, low-cost loan where the interest payments are capitalized and deferred during the 5-year development period. Capitalized interest has been added to the loan quantum and the loan is repaid in full over the remaining term of the loan. And finally, there are no restrictions preventing EVgo from securing additional financing with respect to other portfolios of stalls at any time during the loan term, providing the company with the flexibility to continue to accelerate our growth if attractive financing opportunities exist. Overall, this structure is very flexible and highly efficient. There's effectively no impact on unlevered free cash flows, and we continue to equity finance these stalls over the first 5 years and relatively low-cost debt service payments during the remainder of the loan term. The structure returns the majority of profits to the parent in the form of distributions while a relatively modest restricted cash balance develops over the development period of approximately $16,000 per contributed and newly developed stall in service. A summary of the key loan terms are shown on this slide. As we said, including capitalized interest that is deferred until after the development period, this is a $1.25 billion long-term, low-cost loan provided to a wholly owned project finance SPV. The interest rate is treasury plus approximately 1.2%. Interest is capitalized and amortization deferred during the development period and repaid over the remaining 12-year operational period. The loan and project cash flows will cover up to 100% of gross CapEx, fees and development costs. We borrow 80% of these eligible expenses on a monthly basis, subject to a 65% loan-to-value test, which we don't expect will begin limiting draws until late 2027. Cash flows from the SPV will reimburse EVgo for most of the remainder of eligible expenses. We expect the first draw of over $75 million comprising of $50 million advance and 80% of the eligible costs of around 300 stalls expected to be developed in Q4 this year and Q1 next year that will be funded by the loan and will be received in January 2025. As I said previously, EVgo does not need to contribute any equity. Instead, approximately 1,600 operational stalls will be contributed to the SPV at $235 million original CapEx cost. Cash flows generated by the project are expected to satisfy any additional equity requirements. The result of all this is an attractive and flexible loan structure and terms that allow us to build upon our already successful track record and accelerate the buildout of charging infrastructure with attractive economics for EVgo shareholders. As the largest owner operator of fast charging in the U.S., focused exclusively on building an industry-leading charging network and with a rigorous and proprietary network planning process that is continually back-tested, we've identified over 30,000 charging stalls across the U.S. that meet our return requirements. Because of our scale and the strength of our relationships with our most compelling strategic national and regional site hosts, we've identified specific site hosts for the majority of these potential stalls. And most importantly, there's often more than one site host option for any one location. And for over half of all locations, we have 3 or more site hosts identified. Not only do we believe we have the best locations identified to maximize charging stall performance, but we also have tremendous flexibility to choose between site hosts, allowing us to deploy as fast as we would like. The build plan for DOE-funded fast-charging stalls includes approximately 200 stalls in Q4 2024, rising to a range of around 2,075 to 2,325 by 2029, which is more than 2.5x the number of owned and operated stalls we expect to operationalize this year. This represents 7,500 charging stalls in total. This does not include charging stalls built outside of the DOE Loan Program financing, which include NEVI stalls, EVgo eXtend and dedicated large hubs with many more stalls at our public sites for autonomous vehicle partners. We expect to provide visibility on these areas with our annual guidance. And as I said it earlier, this build schedule assumes gross CapEx per stall we expected going into 2024. If we're able to meet our stated plan to lower gross CapEx per stall by 30%, we would be able to build approximately 1,600 more stalls. Using the midpoint of the annual build schedule for 7,500 stalls and assuming some removals of older stalls through our EVgo ReNew program in 2025, we will more than triple our stalls in operation by 2029 to around 10,600 stalls by the end of that year. The supply-demand picture, together with a series of factors increasing the share of public fast charging, results in an environment supportive of margins at least through the rest of this decade and underpins the growth in throughput per stall in our projected unit economics. There is ample demand for the stalls EVgo expected to deploy under the DOE loan. Demand for DC fast charging is driven by several factors, the most important being EV vehicles in operation or VIO. No matter what scenario was assumed for EV sales through 2030, EV vehicles on the road, or VIO, will increase. The supply of DC fast chargers has not kept up with VIO. EV VIO has grown an annual growth rate of 43% over the last 4 years compared to a much lower annual growth rate of charger supply of 32%. There are currently roughly 49,000 DC fast charging stores from all charging operators across the U.S. And the current operators, only a handful are at scale today, including EVgo, and there's currently a lack of visibility on the growth of aggregate DCFC supply in the future, given the balance sheet requirements. Importantly, EVgo now has the capital and is able to leverage its operational flywheel to build at scale. In addition to the supply of chargers not keeping pace with demand from a VIO standpoint, there are several other factors that we believe will move and has already moved more of the energy required to charge EVs to DCFC networks, resulting in an increase in kilowatt hours per VIO. First is rideshare electrification. Companies such as Uber and Lyft have internal goals to get more drivers to switch to electric, and this is supported by policies requiring rideshare become fully electric in large cities, such as New York City. When a rideshare driver needs to charge up during their shift, they'll usually do so on DCFC networks so they can get back on the roads quickly. Second, autonomous vehicles are beginning to hit the roads in several market pilots from a few companies. The financial use case for AV requires them to be highly utilized. Therefore, when AVs need to charge, they'll use fast charging. EVgo already has partnerships with leading AV firms and has a few dedicated hub sites in operation. Post election, we expect this to be an area of growth that may occur faster than previously thought. Third, as EV adoption moves from early adopter to the mass market driven by more affordable vehicles, more EV drivers are expected to live in multifamily housing without access to home charging. As we've detailed in the past, multifamily EV drivers charge 2x more in our network than single-family EV drivers. Fourth, as vehicles increase their charge rate or the speed at which they take electrons from chargers, it will make the use case for DC fast charging more compelling to drivers. And finally, the standardization of the charging cables to J3400, commonly referred to as NACS, is an opportunity for EVgo. Today, only a small percentage of drivers that use our network are Tesla drivers. As we add J3400 stalls to our network, we are in a unique position to attract roughly 60% of EV VIO to our network that isn't currently using our network today. As I've mentioned before, EVgo stations tend to be in urban, suburban areas closer to amenities than many Tesla stations today. The combination of these factors support our projected unit economics. EVgo is a leader in financial transparency for charging operators. Earlier in 2024, we debuted our per-stall unit economics to help you understand our business and why we believe it is so compelling. We previously estimated that at the scale of approximately 7,000 stalls in 3 to 5 years' time, revenue per stall was estimated to be $91,000 annually and charging network gross profit per stall was estimated at nearly $45,000 annually. With the DOE guaranteed loan financing, we will reach 11,000 stalls between 2029 and 2030, if not earlier, if we're able to lower gross CapEx per stall in line with our plans. We have revised upwards our throughput per stall expectations using the same charge rate assumption and a higher range for utilization, given the growth trajectory we are currently experiencing. This forecast takes into account a lower EV penetration rate assumption for 2030 that might have been expected before the election, but still higher than today and broadly in line with more conservative forecast for EV penetration. We expect charging network gross profit per stall to be $49,500 annually, at the midpoint of our range, which is higher than previously expected. As a reminder, we have tremendous operating leverage in charging network gross margin because around 35% of our current charging network cost of sales is fixed, resulting in higher charging network gross margin with higher throughput per stall as clearly demonstrated from the Q3 2024 annualized actuals. In addition to this operating leverage, we expect to be able to extract further efficiencies at a scale of 11,000 stalls for more sophisticated energy cost management and further economies of scale in operations and maintenance costs within cost of sales. We remain as confident as ever in achieving adjusted EBITDA breakeven in 2025, laying the foundation for a larger and profitable EVgo in the future. Taking the unit economics shown earlier,and assuming a network size of 11,000 stalls, you have a very attractive owned and operated business in several years. And the math is simple. 11,000 stalls, the low and high annual range from the unit economics from the prior slide results in revenue of roughly $1 billion, which is a 7x increase compared to the trailing 12 months through Q3 2024. At 50% to 52% charging network gross margin, charging network gross profit would be approximately $495 million to $594 million. Charging network gross profit growth at 11,000 stalls is an impressive 11x, leveraging the fixed cost of the network within charging network gross margin. We're assuming total adjusted G&A increases up to 2x as we add to our growth G&A to build out the network from the $104 million trailing 12 months through Q3 2024, which again demonstrates the operating leverage in this business. Total adjusted EBITDA is estimated to be $300 million to $425 million when we get to an 11,000 stall network. As a reminder, this is total adjusted EBITDA assuming 11,000 stalls, excluding stalls built outside the DOE loan and excluding the contribution from any other lines of business. This $1.25 billion guaranteed loan facility from the DOE is transformational to EVgo's value as a company. First, there's ample opportunity to grow EVgo's market share in an environment of growing demand but with the DCFC charging supply landscape that is unlikely to grow very fast whereas we've built the growth engine and now have the balance sheet to capitalize on these opportunities. The growth allows us to achieve significant economies of scale, resulting in attractive per-stall unit economics. The large quantum of nondilutive capital allows us to accelerate our rate of growth to achieve these scale benefits and continue growing at an accelerated rate with cash from operations beyond the development period alone. The addition of low-cost debt lowers our weighted average cost of capital, enhancing the value of future cash flows. The net result is a business that compares favorably to high cash flow, high-growth infrastructure peers with higher multiples. As we said up front, accelerating the buildout of charging infrastructure plays a vital role in securing the investments and long-term competitiveness of the U.S. auto industry. In summary, we are delighted to have achieved this important and strategic milestone. This loan builds on our already successful track record of scaling DC fast-charging infrastructure, provides the company with attractive and flexible financing to accelerate our deployments, increase our medium-term adjusted EBITDA expectations and unlocks further shareholder value. With that, operator, we can turn the call over to questions.

[Operator Instructions] Your first question comes from the line of Gabe Daoud from TD Cowen.

Congrats on closing the loan. I was hoping we could maybe just go back to the assumptions a little bit more, Badar. I guess I'm just most curious on -- and again, you laid all this out pretty clearly. But the charge rate going from 50 kilowatts to 80, could you maybe just talk a little bit about that cadence there and, I guess, key factors in driving that? I know you're obviously deploying 350s and above even maybe down the line and then obviously charge rate acceptance on the vehicle side to continue to increase, but maybe just a little more meat on that bone for us would be helpful, I think.

Gabe, yes. It's really what we -- it's exactly what you just said. So we're deploying fast chargers. As you know, the 350-kilowatt chargers are becoming a more -- a bigger part of our network, and that's -- we continue to do that and potentially higher with our next-generation architecture. And as you said already, the VIO mix of the batteries in the cars are just getting faster. You put the 2 things together and you're just going to naturally see an increase. It's a very compelling tailwind for the business. If you go back to couple of years, 2.5 years, that was in the low 30s to an average of 49 today. And so we're seeing more vehicles with faster charge rates coming out, and that's -- the combination of those 2 things get us to the 80 charge rate. That's not an assumption that we've changed from our previous view of our unit economics.

Okay, okay. That's great. Cool. Okay. And then could you maybe just talk to your operational confidence, I guess, in deploying this many stalls annually? It's a pretty significant step up over time versus your historical pace. So maybe a little bit of color on that and maybe an update on some of the known bottlenecks around permitting and transformer shortages and just how deep in planning are you, I guess, in getting this build plan off the ground?

Yes. I mean I think the first thing is do we have the sites, and as we said, we've identified about over 30,000 sites that pencil and meet our return requirements. And across that -- the 30,000 stalls, across those 30,000 stalls, we've -- over half, we've identified 3-plus sites. So we have a pretty sophisticated network plan that looks out over multiple years, and we use that in geographies across the United States to determine where we should build. That gives us the optionality that frankly other companies that don't have this kind of scale can deploy. So if we find that there's a distribution congestion issue at -- on someone utilities network, we'll look at an alternative site. And that's really how we manage that optionality, and that gives us the confidence that we can build and meet the targets as we've done, frankly, all year. I think it helps, the fact that both Paul and I come from utility companies which we ran. So we have an idea of how the utility world works. In terms of scaling up, we built this flywheel. It's been operational for over a decade, Gabe, as you know, and so we've really fine-tuned our processes. We do expect to be growing certain parts of the business, but we'll do that in a very prudent and measured way, which is why you're not seeing a giant increase in our stall count over the near term. But as we go out over the 5 years, we feel pretty confident. And again, I think that there is another operator in the United States that's been operating at this level and above, and so it's -- we're not breaking new ground here.

Okay. Excellent. That's great to hear, and that's good color. And I'll let someone else on. But just a quick clarification. So you borrow 80%, you get essentially most of that 20% back monthly once the stalls become operational, and you think you get your first distribution from the SPV in the first half of '26. Is that right?

That's right. We're contributing about 1,600 stalls, and so the cash flows from those stalls are within the SPV. But that's essentially right. So we expect to get distributions from the SPV back into the sponsor entity in the Q1 2026.

Your next question comes from the line of Craig Irwin from ROTH Capital.

First up, 200 stalls for the DOE build plan in the fourth quarter of '24 and then, I guess, the $75 million disbursement in January of next year seems to be actually a little bit faster than what we were thinking this would start out of the gates. Can you maybe talk a little bit about how this impacts the financial outlook for 2025? I know you're not ready to give official guidance for 2025. But does this maybe bend the curve and make it a little bit easier for you to achieve your goal of EBITDA profitability during the next year?

Craig, yes, we -- no, look, we're obviously very pleased about being able to contribute the 200 stalls in from the fourth quarter of this year, and the advance is $50 million plus $25 million for 300 stalls for both Q4 and [indiscernible] on. So we're pleased with that. I can't -- I'm not going to give any guidance, an additional guidance for 2025, Craig. But other than to say, as I said on the call, we remain fully committed and expect to be able to be EBITDA breakeven in 2025.

Okay. Excellent. My second question is just looking at the broader recalibration around expectations for EV market growth that have been really ongoing for the last year. There's now sort of expectations of tepid but positive growth any of these throughout 2025, where the market leader most likely continues to take share versus other OEMs. Can you maybe talk a little bit about the concentration of NACS-complianced stalls that you might be looking to install this next year? Do you have sufficient access to cables to install a much greater mix of these stalls? What are the priorities as far as NACS compliance on the units that are rolled out over the next couple of years?

Yes. We're very focused, Craig, on the -- on deploying NACS cables onto our network for new stalls and potentially as retrofit for some of our existing stalls. And we've been going through the -- along with the whole industry, the standardization, the J3400 standard, making sure that it's compliant with the various tests. We've got very high-speed chargers as we just talked about. It required liquid cooled cables. And as soon as that is done, we start -- we expect to be able to deploy those to our network. We don't have -- on the -- we'll provide some more color on specific rollout of NACS cables in our Q4 call as part of our annual guidance. But as I said before, and I think as you've asked, we're very excited about the NACS cable because today, a very small percentage of Tesla vehicles are charging in our network despite the fact that most of our network is likely closer to where people live, work and want to -- and go about their errands versus Tesla's network that tends to be more highway-focused. And so being able -- and as you pointed out, Tesla's share of VIO is maybe 60%, potentially growing if they grow share next year. And so getting access to those vehicles is something we're very excited about because it allows us to grow throughput on our network without any increase in VIO, which is amazing.

Okay. Excellent. And then last thing, just as a point of clarification. I think you were pretty clear on this. But the DOE stall build plan for '25, the 750 to 850 units, that's additive to your other opportunities with different avenues of funding. This is a very rapid acceleration at the company and something that can take a material impact in a relatively short period of time. Am I looking at that correctly?

The build schedule that we laid out are the stalls that we will build as a result of the DOE financing. So it does not include any stalls that we build outside of financing, which might include NEVI stalls. It might include dedicated hubs for autonomous vehicle partners. It's small part of our business today, but, as I think many people have said, could be a potential source of significant upside in the short to medium term. But however, I would say, Craig, the majority of the stalls that we'll build in the near term certainly are going to be stalls that are financed through the DOE build program because it's obviously very low-cost capital. So it is the majority of the stalls that we'll build.

Your next question comes from the line of Chris Dendrinos from RBC Capital Markets.

Congratulations. I guess maybe to start here, you laid out that you've got 30,000 identified stalls that meet the payback criteria. So is that 30,000 that meet that criteria today? And then does, I guess, mean that you could, I guess, potentially develop faster given the plethora of opportunities there? Or I guess like what factors go into the development pace that you've laid out today?

Yes. Chris, that's right. That 30,000 is what meets our criteria today. And so as EV adoption grows, we'd expect to see that grow. And in terms of pace, look, at the end of the day it comes down to 2 things: do you have the capital that's attractive, which we've just locked up a giant slug of capital that's attractive for us; and then do we have the operational capacity to actually -- to grow. We think we've built a flywheel that allows us to do that. And we put the 2 together, and we can grow pretty fast. What we will not do is not meet our commitments, run EBITDA breakeven. And so there are costs that we will incur on the income statement, not just capital costs, as we ramp up growth. And so we're going to be measured in our rate of growth, our pace of growth so that we are -- we remain focused on meeting EBITDA breakeven next year. And at that point, we'll have covered our fixed costs. And so all stall-based cash flow, once our fixed costs are covered, goes straight to the bottom line, and you can accelerate the rate of growth faster and still meet EBITDA expectations going forward. So that's really how we think about it. In terms of other opportunities, if it turns out that there are -- it makes sense for us to grow faster, then we'll look to deploy the capital from the DOE faster. And one way of doing that is by lowering the cost -- the capital cost of the stalls. As I said on the call, if we're successful lowering our capital cost with the plans that we've already publicly talked about several times over the course of this year, together with the MOU with Delta, that allow us to lower our capital cost by 30%, you'd see an extra 1,600 stalls in this time period within the same $1.25 billion loan. And if there are other opportunities that allow us to deliver shareholder value even faster, then we would look at potentially exploring other nondilutive financing that we put on top of this.

Got it. Okay. And then I guess maybe as my follow-up. I'm going to sneak 2 into 1 here. If the 30Cs rules change or repealed under the new administration, how does that affect the development plan? And then -- or I guess, does it affect the development plan? And separately, just on the G&A ramp. Is that coincide, I guess, with the ramp in activity? Or could that be more front-end weighted to, I guess, get in front of and have a personnel or operational scale to support that future development, how we'll get there?

Yes. So on the 30C, look, we -- our offsets today -- capital offsets, as we talked about on many earnings calls, are roughly about 50% of our gross capital cost per stall. That's a bit of 30C, some NEVI state grants and the infrastructure payments that we get from GM. We've assumed conservatively offsets of around 25% for this build plan. So we're -- we think we're being quite conservative here and feel pretty comfortable about that. Clearly, if offsets are greater, then it's more cash flow for the company. In terms of G&A, we've kind of broken down G&A at -- on the prior earnings calls: 30% sustained G&A, 30% growth G&A, 40% of our G&A is fixed costs. So it's the 30% that's considered growth G&A that's tied to the growth of our -- the annual rate of growth. So you'd expect to see that grow over the course of the next several years. Clearly, sustaining G&A will rise because that's completely linked to stalls and operation. And in terms of front or back-end loaded, I'd expect to see the majority of the growth in G&A through sustaining G&A, which means when stalls are in operation generating cash, they're covering the sustaining cost of those stalls.

Your next question comes from the line of Bill Peterson from JPMorgan.

Yes. Congrats on the DOE loan close. I had kind of a question on competition and maybe as a follow-up to Chris' question earlier. I can envision now that you've closed this loan, there may be an increased urgency in terms of a land grab maybe potentially from competitors as well, especially given like how EV growth has definitely outpaced out of charger. So it looks like if I look at your host site map, it looks like you have multiple host sites in over 70% of locations. But are there any concerns around competitors getting to your sites that you'd like to build between now and 2029? Or asked in another way, I mean, are you able to reserve some of these sites ahead of them in order to lock in what may be considered to be the best sites?

We really don't have -- hey, Bill. We don't have any concerns on honestly at all. We've got great relationships with site hosts. We've got a ton of optionality here. As I look -- if I look backwards, as we shared in the prior earnings call, we think that we're just getting better and better at where we're selecting sites. If you just look at the -- on the second quarter earnings call, we showed our stall performance by vintage year. And you'll see that the 2023 vintage stalls as of Q2 of this year, we're performing about 75% better than vintage 2020 stalls. So that just is a reflection of we're getting better at figuring out which sites are -- have better stall performance. We've got great relationships with site hosts, and we've got an enormous amount of optionality. That slide that we showed is 30,000 stalls. We're really looking at 7,500 stalls, potentially 1,600 more if we lower -- if we're successful in lowering our CapEx program over the next 5 years. So we just don't have concerns. In terms of the landscape, as you know, there's 30 to 40 different operators. Almost none of them have the level of scale that we're operating at. And so we just don't believe that most of the rest of the market is that sophisticated in site selection as we are.

Okay. Earlier, you talked, and just now you also mentioned, about the performance requirements. Can you elaborate what that means? I guess -- I think it was related to the STD (sic) [ SPV ], but I guess you were saying you need to meet certain performance requirements. But what is that? Is that uptime, availability? And I guess does the performance, as you define it, have to approve -- improve from here? Or are you meeting those performance requirements already?

Yes. Great question. The performance tests are effectively based on stall performance, so kilowatt hours per stall per day, so throughput as we talk about on our calls. And so if I look at that for both draws and distributions during the development period, we expect to continue to receive draws and distributions if there's no increase in stall performance during the first 5-year deployment period, so no increase whatsoever. And again, I think as I answered before, I think it was Craig's question, we expect distributions by Q1 2026. In the operational period, so that's after the first 5 years, the second 12 years on the 17-year term, distributions continue even if stall performance growth -- the growth in stall performance is half the increase to the bottom end of the range in our projected unit economics that we showed in the right-hand column at 450-kilowatt hours per stall per day. So even if our growth was just half of that increase, we would still receive distributions during that second -- the 12-year operational period. So that's a pretty wide margin of safety, we feel. And as you know -- as you've seen, I mean, our kilowatt hours per stall per day, our throughput continues to grow. It was a 12% growth this past quarter. And as long as 2 things were happening: VIO growth is above DCFC supply, which it has been and expect to be over the next decade; and the share of kilowatt hours increases as a percentage of total kilowatt hours in terms of home, workplace or public charging, we expect that, that growth in throughput per stall is pretty secure.

Your next question comes from the line of Stephen Gengaro from Stifel.

So a couple of things for me. I think -- why don't I start with this, can you just -- just talking about CapEx per stall, can you just give us kind of the current average level that you're talking about on kind of gross CapEx per stall? And how is the supply chain to meet kind of the expansion that you're laying out here?

Yes. We -- so we began the year -- Stephen, we began the year with an expectation that gross CapEx per stall would be around $160,000. And as I said on the Q3 earnings call just a month ago, we expect that actually -- that's coming in this year about 6% lower. We expect 2025 -- that's for 2024 vintage stalls, in other words, stalls that go operational in 2024. For 2025 vintage stalls, we're expecting that to be about 10% lower from that $160,000-ish level. That's gross CapEx per stall. That's not included in the offsets. Again, this year, we're -- offsets are around 50%. And the program that we've been talking about all year for next-generation architecture, we're targeting a 30% reduction in gross CapEx per stall, again, from that baseline of around $160,000. And that'll be for stalls that we start deploying in the second half of 2026. The partnership, the MOU that we announced with Delta Electronics is what that's really all about. So we're bringing our experience and customer pain points together with Delta's experience in global power electronics together to build a new charging architecture, and so we're very excited by that. And I don't expect any supply chain issues. Quite honestly, we -- especially on the back of this loan -- as well as on the back of this loan, and we've got great relationships already with supply chain. And I think this sets us up even better in terms of managing those relationships, securing new relationships that necessary as well. And again, if we are able to -- if we're successful lowering the CapEx cost per stall, as I said, at that 30% level, you would be looking at an extra 1,600 stalls within the loan financing from the DOE.

Great. That's good color. And then the other one -- now we can take some of this off-line. But you talked about the special purpose vehicle and kind of laid that out with a lot of detail. But like optically for us on the income statement, what is this going to look like? Is it going to look any different? I mean are there going to be some -- or is it going to show up kind of how we've been seeing it show up on the income statement, on the cash flow statement?

At the end of the day, it doesn't -- it's you don't need to think about it any differently than we are today. Unlevered free cash flow, the same in the first 5 years. We have no debt service payments for the first 5 years until the operational period, right? So interest is capitalized and deferred. We get $193 million of capitalized interest. So there's no -- there's really no impact on cash flows in the first 5 years. It's the second 12 years we've got these relatively low-cost debt service payments that you'll need to put in the income statement.

Yes. And Badar, if I could add -- great. Wonder if I could just add that -- we'll still be talking about the business holistically. You'll see the EVgo, Inc. results in metrics, and sort of the SPV, of course, is subsumed within that as well. So really, it would be just as transparent as it's been in the cost.

Yes. Thanks, Paul.

Your next question comes from the line of Doug Becker from Capital One.

Congratulations. Maybe just an add-on to the last questions about the income statement. How will we see it showing up on the balance sheet, just from a high level?

Yes. Paul, do you want to take that again as well?

Yes. I mean like I was just saying, you'll see on the balance sheet, the consolidated balance sheet, which will show the loan and it'll show the reserve accounts as well that need to be funded as part of the loan and unrestricted cash as well. So just exactly what you would expect to see for any other kind of project financing.

Got it. And then I appreciate we'll get formal guidance on the stall growth outside the DOE loan next year. In the interim, is it just best to assume something materially lower than the 800 to 900 stall installation pace that had been laid out previously? And I guess I'm just trying to think just a couple of hundred of non-DOE stalls as we think about 2025.

Yes. I think that -- I think for 2025, I think you should assume that the DOE build plan is likely the majority of the build plan. And again, the -- what we're talking about here, these are our owned and operated stalls. So we're not including the eXtend buildout. That's obviously not -- that's not a owned stall deployment. We operate the stalls for eXtend but we don't own them. So the stuff that's outside of the DOE build plan would be -- that's owned and operated would be NEVI stalls. It might be dedicated hubs for the autonomous vehicle partners, again not a huge part of our business today, but likely could be a big area of growth. So I would say that what we're showing here, at least for the first year, is likely to be the majority of our build.

Got it. And just a real quick one. If, let's say, 7,500 stalls aren't ultimately delivered in the 5-year period, are there any penalties? Or is there anything looking further out that could be a negative?

No.

Your next question comes from the line of William Grippin from UBS.

Great. Badar, just wanted to come back to a couple of comments you made earlier. It sounds like you're expecting a majority of the obligation under the GM partnership to now fall within this DOE structure as well. So if I think about that in combination with potentially half of the sites being eligible for 30C, should we be thinking about some site deployments actually creating net cash to EVgo, meaning you're not actually using any of your own equity to build out these sites?

Will, so yes, the GM stalls will be covered by the DOE loan, and yes, there -- we will be receiving offsets like 30C or potentially other grants on top of what we get -- what we borrow from the DOE. So that's correct. As I said, this is a attractive and flexible structure for us.

Yes, indeed. And then just to clarify on the performance tests, that is specifically related to the stalls you're contributing as collateral, correct?

No. The stall -- the performance tests for draws -- well, draws are for new stalls, obviously. But the performance test for both draws and distribution are for both the 1,600 stalls that we contribute as well as the stalls that we develop within the SPV.

And that concludes our question-and-answer session. I will now turn the call back over to Badar Khan for closing remarks.

Great. Well, thank you. As we said upfront, we're delighted to have reached this pretty important and strategic milestone. It builds on our track -- very successful track record of scaling up the business to date. It provides us with a attractive and flexible financing that allows us to accelerate our deployments, increase our medium-term adjusted EBITDA expectations and I think, as a result, unlocks very strong shareholder value. So thank you for joining the call, and look forward to talking to you all next time. Thanks very much.

This concludes today's conference call. Thank you for your participation. You may now disconnect.